Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 14, 2020

• Fix torch.cuda.check_error type errors (pytorch#41330)

  Show description for c20426f
  

  lw

  authored and

  facebook-github-bot

  committed
  c20426f

  Copy full SHA for c20426f

• Add torch.logit function (pytorch#41062)

  Show description for 80d5b37
  

  xiaomengy

  authored and

  facebook-github-bot

  committed
  80d5b37

  Copy full SHA for 80d5b37

• Remove thrust casting from static_cast_with_inter_type (pytorch#39905)

  Show description for 34e11b4
  

  zasdfgbnm

  authored and

  facebook-github-bot

  committed
  34e11b4

  Copy full SHA for 34e11b4

• Fix FC issue (pytorch#41198)

  Show description for 5f6c6ed
  

  smessmer

  authored and

  facebook-github-bot

  committed
  5f6c6ed

  Copy full SHA for 5f6c6ed

Commits on Jul 13, 2020

• move misc implementation out of jit/__init__.py (pytorch#41154)

  Show description for ca1b8eb
  

  suo

  authored and

  facebook-github-bot

  committed
  ca1b8eb

  Copy full SHA for ca1b8eb

• add spaces in .md annotation for python indent (pytorch#41260)

  Show description for 6392713
  

  蔡舒起

  authored and

  facebook-github-bot

  committed
  6392713

  Copy full SHA for 6392713

• .circleci: Explicitly remove nvidia apt repos (pytorch#41367)

  Show description for b6e1944
  

  seemethere

  authored and

  facebook-github-bot

  committed
  b6e1944

  Copy full SHA for b6e1944

• update operators in the mapping to fp16 emulation

  Show description for d601325
  

  Hector Yuen

  authored and

  facebook-github-bot

  committed
  d601325

  Copy full SHA for d601325

• helper function to print out all DDP-relevant env vars (pytorch#41297)

  Show description for 4196605
  

  Yi Huang (PyTorch)

  authored and

  facebook-github-bot

  committed
  4196605

  Copy full SHA for 4196605

• fix duplicate extern sdot and missing flags (pytorch#41195)

  Show description for 6e6931e
  

  EscapeZero

  authored and

  facebook-github-bot

  committed
  6e6931e

  Copy full SHA for 6e6931e

• Quantization: preserving pre and post forward hooks (pytorch#37233)

  Show description for 0c77bd7
  

  zetyquickly

  authored and

  facebook-github-bot

  committed
  0c77bd7

  Copy full SHA for 0c77bd7

• Add shape inference functions for int8 quantization related ops (pytorch#41215)

  Show description for c451dda
  

  csummersea

  authored and

  facebook-github-bot

  committed
  c451dda

  Copy full SHA for c451dda

• Add interpolate-style overloads to aten::upsample* ops (pytorch#37176)

  Show description for 7183fd2
  

  dreiss

  authored and

  facebook-github-bot

  committed
  7183fd2

  Copy full SHA for 7183fd2

• Add support for float[]? arguments in native_functions.yaml (pytorch#37175)

  Show description for fb9e44f
  

  dreiss

  authored and

  facebook-github-bot

  committed
  fb9e44f

  Copy full SHA for fb9e44f

• Back out "Revert D22329069: Self binning histogram" (pytorch#41313)

  Show description for d04a2e4
  

  Yavuz Yetim

  authored and

  facebook-github-bot

  committed
  d04a2e4

  Copy full SHA for d04a2e4

• .cirlceci: Setup nvidia runtime for cu as well (pytorch#41268)

  Show description for 86d803a
  

  seemethere

  authored and

  facebook-github-bot

  committed
  86d803a

  Copy full SHA for 86d803a

• reduce logging for layernorm (pytorch#41305)

  Show description for dea39b5
  

  Hector Yuen

  authored and

  facebook-github-bot

  committed
  dea39b5

  Copy full SHA for dea39b5

• Pass the number of indices but not embedding size in PyTorch operator (pytorch#41315)

  Show description for 67a4f37
  

  Mengchi Zhang

  authored and

  facebook-github-bot

  committed
  67a4f37

  Copy full SHA for 67a4f37

• Impl for ParameterList (pytorch#41259)

  Show description for 98df978
  

  arktrail

  authored and

  facebook-github-bot

  committed
  98df978

  Copy full SHA for 98df978

Commits on Jul 12, 2020

• Fake Quantization Per Channel Kernel Core Implementation (CPU) (pytorch#41037)

  Show description for fa15318
  

  Paul Shao

  authored and

  facebook-github-bot

  committed
  fa15318

  Copy full SHA for fa15318

• Fake Quantization Per Tensor Kernel Core Implementation (CPU) (pytorch#41029)

  Show description for 5e72ebe
  

  Paul Shao

  authored and

  facebook-github-bot

  committed
  5e72ebe

  Copy full SHA for 5e72ebe

• [quant] Adding zero point type check for per channel quantization (pytorch#40811)

  Show description for 402be85
  

  z-a-f

  authored and

  facebook-github-bot

  committed
  402be85

  Copy full SHA for 402be85

Commits on Jul 11, 2020

• [quant][graphmode] use RemoveMutation to remove append (pytorch#41161)

  Show description for 4b4184f
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  4b4184f

  Copy full SHA for 4b4184f

• Op to create quant scheme blob (pytorch#40760)

  Show description for 106b0b6
  

  Anurag Gupta

  authored and

  facebook-github-bot

  committed
  106b0b6

  Copy full SHA for 106b0b6

• [quant] dequantize support list and tuple of tensors (pytorch#41079)

  Show description for edcf2cd
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  edcf2cd

  Copy full SHA for edcf2cd

• Add fp16 support to SparseLengthSum PyTorch operator (pytorch#41058)

  Show description for c864158
  

  Mengchi Zhang

  authored and

  facebook-github-bot

  committed
  c864158

  Copy full SHA for c864158

• [caffe2] Revert D22220798 (pytorch#41302)

  Show description for 28291d3
  

  Hao Lu

  authored and

  facebook-github-bot

  committed
  28291d3

  Copy full SHA for 28291d3

• fix the range of the random weights used in the int8fc test (pytorch#41303)

  Show description for e544bf2
  

  Hector Yuen

  authored and

  facebook-github-bot

  committed
  e544bf2

  Copy full SHA for e544bf2

• Revert D22467871: add check for duplicated op registration in JIT

  Show description for a1ed6e1
  

  linbinyu

  authored and

  facebook-github-bot

  committed
  a1ed6e1

  Copy full SHA for a1ed6e1

• [caffe2] Fix the issues when using CUB RadixSort (pytorch#41299)

  Show description for 095886f
  

  jianyuh

  authored and

  facebook-github-bot

  committed
  095886f

  Copy full SHA for 095886f

• Solve log2(x:int) ambiguity by using log2(float(x)) (pytorch#41295)

  Show description for d1f06da
  

  malfet

  authored and

  facebook-github-bot

  committed
  d1f06da

  Copy full SHA for d1f06da

• Fix arg type annotations in jit.trace and onnx.export (pytorch#41093)

  Show description for 1c098ae
  

  JBamberger

  authored and

  facebook-github-bot

  committed
  1c098ae

  Copy full SHA for 1c098ae

• Ampere has CUDA_MAX_THREADS_PER_SM == 2048 (pytorch#41138)

  Show description for 877a599
  

  zasdfgbnm

  authored and

  facebook-github-bot

  committed
  877a599

  Copy full SHA for 877a599

• Better THGeneric.h generation rules in bazel (pytorch#41285)

  Show description for 6cbb924
  

  malfet

  authored and

  facebook-github-bot

  committed
  6cbb924

  Copy full SHA for 6cbb924

• Revert D22465221: [pytorch][PR] Reducing size of docker Linux image

  Show description for 67f5d68
  

  malfet

  authored and

  facebook-github-bot

  committed
  67f5d68

  Copy full SHA for 67f5d68

Pagination

Previous

Next